A magnet can steer a moving charge without ever speeding it up or slowing it down. That one odd fact is how laboratories sort atoms by weight, how the Large Hadron Collider keeps protons in a ring 27 km round, and why the aurora hangs over the poles. The other half of the story is where magnetic fields come from. They're made by charges that move. Magnetism is what electricity does when it's in motion.
The magnetic field is written \(\mathbf{B}\). Like the electric field it has a size and a direction at every point, and the first figure looks straight along it.
- Watch a lap. The amber force is always at right angles to the white velocity, pointing at the centre of the circle. The speed trace in the corner stays level at 100.0 km/s.
- Drag the amber point in until the launch is 50 km/s. The circle is half the size, 3.48 cm, and the readout says a lap takes exactly as long as before.
- Go back to 100 km/s and raise the field to 30 mT. Half the radius again, and this time half the lap time too. Then put it back to 15 mT.
- Make the charge −1 e. It circles the other way. Then +2 e: twice the force, half the radius. Put it back to +1 e.
- Double the mass. The circle doubles, to 13.92 cm, and swings out of the picture before it comes back: heavier things are harder to steer. Put it back to one proton mass.
- Slide the field through zero to "out of the page". The crosses become dots, arrows coming towards you, and everything turns the other way.
- Set the charge to zero.
- Now set 15 mT into the page, +1 e, one proton mass and 100 km/s straight across, and turn the electric field up to 1.5 kV/m down the page. The particle flies straight. Shorten the launch to 60 km/s and it curls downward. Lengthen it to 140 km/s and it curls up.
The force on a moving charge
A charge \(q\) moving with velocity \(\mathbf{v}\) through a magnetic field feels a force
The \(\times\) is the cross product from dot and cross products, the same one that gives a torque its direction. It makes a vector at right angles to both \(\mathbf{v}\) and \(\mathbf{B}\), with size
where \(\theta\) is the angle between the velocity and the field. For the direction, use your right hand: point your fingers along \(\mathbf{v}\), curl them round towards \(\mathbf{B}\), and your thumb points along the force on a positive charge. A negative charge is pushed the opposite way. Try it on the figure: fingers to the right, curling into the page, and the thumb points up the page.
The field is measured in tesla, T. One tesla puts a force of 1 N on a charge of 1 C moving at 1 m/s across it. That's a strong field. The Earth's is about 0.00005 T (50 µT), a fridge magnet's a few thousandths of a tesla (a few mT), and a hospital scanner's 1.5 or 3 T.
Three things follow from the formula.
- A charge at rest feels nothing. With \(v = 0\) there's no force, however strong the field.
- So does one moving along the field. Then \(\theta = 0\) and \(\sin\theta = 0\).
- The force does no work. It's always at right angles to the motion, and only the part of a force along the motion does work. So a magnetic field can turn a charge but can never change its speed or its kinetic energy.
Why a circle
A force of steady size, always at right angles to the velocity, is what a circular orbit needs. There the pull was gravity. Here it's \(qvB\), for a charge moving square across the field (\(\sin\theta = 1\)).
Something moving at speed \(v\) round a circle of radius \(r\) is accelerating towards the centre at \(\frac{v^2}{r}\). The magnetic force is what provides it:
Divide both sides by \(v\), then rearrange for \(r\):
One lap is \(2\pi r\), covered at speed \(v\). Put in the \(r\) just found, and the \(v\) cancels:
Faster or heavier particles swing wider, and a stronger field or a bigger charge pulls them in tighter. For the figure's proton (\(m = 1.67 \times 10^{-27}\) kg, \(q = 1.60 \times 10^{-19}\) C) at 100 km/s in 15 mT:
The lap time has no \(v\) in it. A faster particle has further to go and covers it in the same time. Ernest Lawrence built the cyclotron on that fact in the 1930s: give the particles a push twice every lap, on a fixed beat, and they stay in step as they speed up, until they near the speed of light.
The radius sorts particles. Send atoms with the same charge and speed through the same field and the heavier ones swing wider, so they land in different places. That's a mass spectrometer, and it's how the isotopes of ordinary elements such as neon were first told apart.
Crossed fields
An electric field pushes on a charge whether it's moving or not. With both fields present the forces add:
In the last thing you tried, the electric force \(qE\) pointed down the page and the magnetic force \(qvB\) up it. They cancel when
The charge has cancelled, and the mass never appeared. Any charged particle moving at \(\frac{E}{B}\) goes straight through, and everything faster or slower is turned aside. With 1.5 kV/m and 15 mT that's \(\frac{1500}{0.015} = 100{,}000\) m/s. In 1897 J. J. Thomson used crossed fields like these to find the speed of the particles in a cathode ray, and from it their ratio of charge to mass. They were electrons.
A slanted launch
Fire the charge at an angle to the field and split its velocity into components. The part along the field gives no force, so it carries on unchanged. The part across the field is bent into a circle. Put together, the charge winds along the field line in a helix.
- Drag the amber point down until the arrow lies along the field. No force at all: a straight line.
- Drag it straight up, square to the field. Now it circles on the spot, as in the first figure.
- In between, tilt it back and forth. More of the velocity across the field makes the coil wider. More along it stretches the coil out.
- Raise the field. The coil gets tighter and shorter together, because the radius and the lap time both go as \(\frac{1}{B}\).
This is the aurora. Charged particles, mostly electrons, that the Sun's wind has fed into the Earth's magnetic field spiral along its field lines. Near the poles the lines plunge down into the atmosphere, and the particles follow them in and strike the air about 100 km up. Oxygen glows green and red, nitrogen blue and purple.
Currents make magnetic fields
In 1820 Hans Christian Ørsted noticed a compass needle twitch when he switched on a current in a wire nearby. Until then electricity and magnetism had been two subjects.
- Drag the compass right round the wire, keeping the same distance. It always points along the circle, and the field is the same strength all the way round.
- Drag it out from 4.0 cm to 8.0 cm. The field halves, from 50.0 to 25.0 µT. Bring it in to 2.0 cm and it's 100.0 µT.
- Press "Reverse the current". Every needle swings round to point the other way.
- Turn the current down to 2 A. At 4.0 cm the field is 10.0 µT, a fifth of the Earth's. That's why ordinary wiring doesn't upset a compass much.
The field lines are circles round the wire. For their direction there's a second right-hand rule: grip the wire with your thumb along the current, and your fingers curl the way the field goes. For the size, there's a rule named after André-Marie Ampère. Go once round any closed loop, adding up the part of \(\mathbf{B}\) that points along each step (a line integral). The total is a constant, \(\mu_0\), times the current passing through the loop.
Take the loop to be a circle of radius \(r\) centred on the wire. Nothing about the wire picks out one side of it from another, so \(B\) is the same all the way round, and it points along the circle. Adding it up round the loop is just \(B\) times the circumference:
Divide by \(2\pi r\):
With that value of \(\mu_0\) the formula is \(B = 2 \times 10^{-7} \times \frac{I}{r}\). It falls off as \(\frac{1}{r}\): twice as far away, half the field.
Fields from different currents add, as arrows. Wind the wire into a long coil, a solenoid, and the fields of all its turns add up inside it into one strong, even field: \(\mu_0\) times the current times the number of turns in each metre. A coil with 1,000 turns to the metre carrying 12 A gives the first figure's 15 mT. A permanent magnet gets its field from its electrons, as the next lesson explains.
Two wires
A current is charge on the move, so a wire carrying one feels a force in a magnetic field. In a length \(L\) of wire with cross-section \(A\) there are \(nAL\) free electrons, where \(n\) is how many there are in each cubic metre. Each has charge \(e\) and drifts at speed \(v\), so if the field is square to the wire each feels a force \(evB\). Add them up:
using \(I = nAve\) from the circuits lesson. Now put a second wire beside the first. Each one sits in the other's field.
- With both currents towards you, the wires pull together. Check one with your right hand: on the right-hand wire, fingers towards you along the current, curling up the page towards the blue arrow. Your thumb points left.
- Press "Reverse the right wire". Now they push apart.
- Put the currents the same way again and drag the right wire in from 6.0 cm to 3.0 cm. The force doubles, from 333 to 667 µN on each metre.
- Double one current, then both.
- Set one current to zero. The other still makes a field, but there's nothing for it to push on.
Call the currents \(I_1\) and \(I_2\) and the gap between the wires \(d\). At the second wire, the first one's field is
It's square to the second wire, so the force on a length \(L\) of that wire is \(I_2LB_1\):
Divide by \(L\) for the force on each metre:
Currents the same way attract. Currents opposite ways repel.
Until 2019 this was how the amp was defined. One amp was the current that, in each of two long wires a metre apart, gives a force of exactly \(2 \times 10^{-7}\) N on every metre. That made \(\mu_0\) exactly \(4\pi \times 10^{-7}\). Since 2019 the amp has been defined by fixing the charge of the electron instead, and \(\mu_0\) is something to measure. It agrees with the old value to better than one part in a billion.
A charge that's sitting still feels only an electric field, and one that's moving feels a magnetic one as well. But whether a charge is moving depends on who is watching. That puzzle was one of the things that led Einstein to special relativity, in which electric and magnetic fields turn out to be two views of one thing. And look again at the field lines round the wire. They are closed circles, with no start and no end, which the height map of the voltage lesson can't produce. The next lesson finds the kind of potential that can.
Problems
An electron (charge of size \(1.60 \times 10^{-19}\) C) moves at \(2.0 \times 10^6\) m/s at right angles to a field of 0.010 T. What force does it feel? What if it's moving along the field instead? Compare the first answer with the electron's weight (\(m = 9.11 \times 10^{-31}\) kg).
Answer
\(F = qvB = 1.60 \times 10^{-19} \times 2.0 \times 10^6 \times 0.010 = 3.2 \times 10^{-15}\) N. Along the field, \(\sin\theta = 0\) and there's no force. Its weight is \(mg = 9.11 \times 10^{-31} \times 9.81 \approx 8.9 \times 10^{-30}\) N, which is about \(3.6 \times 10^{14}\) times smaller. That's why the figures can leave gravity out.
A proton (\(m = 1.67 \times 10^{-27}\) kg, \(q = 1.60 \times 10^{-19}\) C) moves at 100 km/s at right angles to a field of 15 mT. Find the radius of its circle and the time for one lap. What are they at 50 km/s? Check all four on the first figure.
Answer
\(r = \frac{mv}{qB} = \frac{1.67 \times 10^{-27} \times 1.0 \times 10^5}{1.60 \times 10^{-19} \times 0.015} = \frac{1.67 \times 10^{-22}}{2.40 \times 10^{-21}} \approx 0.0696\) m, or 6.96 cm. \(t_{\text{lap}} = \frac{2\pi m}{qB} = \frac{2\pi \times 1.67 \times 10^{-27}}{2.40 \times 10^{-21}} \approx 4.37 \times 10^{-6}\) s. At half the speed the radius halves to 3.48 cm, and the lap time doesn't change.
A cyclotron uses a field of 1.5 T. How many laps a second does a proton make in it? What's the radius of its path when it has reached \(3.0 \times 10^7\) m/s, a tenth of the speed of light?
Answer
\(t_{\text{lap}} = \frac{2\pi m}{qB} = \frac{2\pi \times 1.67 \times 10^{-27}}{1.60 \times 10^{-19} \times 1.5} \approx 4.37 \times 10^{-8}\) s, so it makes \(\frac{1}{4.37 \times 10^{-8}} \approx 2.3 \times 10^7\) laps a second, at any speed well below light's. \(r = \frac{mv}{qB} = \frac{1.67 \times 10^{-27} \times 3.0 \times 10^7}{1.60 \times 10^{-19} \times 1.5} \approx 0.21\) m.
You want a velocity selector that lets through particles moving at 100 km/s, using a magnetic field of 15 mT. What electric field do you need? If it's made by two plates 1.0 cm apart, what voltage goes across them? Which way does a proton at 140 km/s get pushed?
Answer
\(E = vB = 1.0 \times 10^5 \times 0.015 = 1{,}500\) V/m. From the voltage lesson, \(E = \frac{V}{d}\), so \(V = Ed = 1500 \times 0.010 = 15\) V. At 140 km/s the magnetic force \(qvB\) is bigger than the electric force \(qE\), so the proton is pushed the way the magnetic force points.
A long wire carries 10 A. What's the field 4.0 cm from it? A second wire, parallel to the first and 4.0 cm away, also carries 10 A. What force acts on each metre of it? Then find the force on each metre of two wires 1 m apart, each carrying 1 A.
Answer
\(B = 2 \times 10^{-7} \times \frac{10}{0.040} = 5.0 \times 10^{-5}\) T, which is 50 µT. On one metre of the second wire, \(F = ILB = 10 \times 1 \times 5.0 \times 10^{-5} = 5.0 \times 10^{-4}\) N, or 500 µN: about the weight of two grains of rice. For 1 A in each, a metre apart, \(\frac{F}{L} = 2 \times 10^{-7} \times \frac{1 \times 1}{1} = 2 \times 10^{-7}\) N, the old definition of the amp.
In the second figure as it first appears, a proton is fired at 80 km/s, at \(70^\circ\) to a field of 15 mT. Find the parts of its velocity along and across the field, the radius of its helix, and how far it moves along the field in one turn. Use \(\sin 70^\circ = 0.940\), \(\cos 70^\circ = 0.342\) and the lap time from the second problem.
Answer
Along the field: \(80 \times 0.342 \approx 27.4\) km/s. Across it: \(80 \times 0.940 = 75.2\) km/s. Only the part across the field goes into the radius: \(r = \frac{1.67 \times 10^{-27} \times 7.52 \times 10^4}{2.40 \times 10^{-21}} \approx 0.0523\) m, or 5.23 cm. A turn takes \(4.37 \times 10^{-6}\) s whatever the speed, and in that time the proton moves \(2.74 \times 10^4 \times 4.37 \times 10^{-6} \approx 0.120\) m along the field: 12.0 cm.